Give Me The Rock
this is fantasy basketball
Here comes a blockquote courtesy of an email I got a while ago…
Hey guys – I’ve been a lurker at your site for a while. Great site!
I was writing in because Iwanted to see if any of you would be willing to think through something with me.
I know you have your rankings, and I have something similar. One thing I’ve been working on, though, is ranking players for draft purposes.
I’ll tell you one thing I’m doing different – I’m ranking players, but then re-ranking those players based on the depth at any given position. Sounds crazy, maybe, but my thinking is this: suppose you are in a 10 team fantasy draft and you have your pick between Amare Stoudemire and Kobe Bryant. Also, suppose it’s just prior to this past season and everyone magically had perfect knowledge of what the next year’s stats would be. Who would you take in your fantasy draft? In the rankings you guys compile you have them pretty close, but the casual fan would probably think Bryant is the “obvious” choice.
Getting to a preliminary ranking is the first trick, but then I want to consider the fact that in a 10 team league, SOMEBODY is going to have to start the 10th best center and 10th best SG (or worse). So if I pass were to pass on a center, I might have to throw out some scrub to complete my lineup if centers are scarce. It’s possible SG might be pretty deep, so not getting Bryant might not be the end of the world.
You can go too far with this, though – you can’t draft JUST by positions. So I’ve been using this as just one piece of the puzzle.
I wanted to see if I can get one of you to sanity-check me and offer feedback.
Here was my response:
Obviously getting an idea of depth at a position is an important part of ranking players. In the case you mention, though, I’m still going with Kobe. I hope that’s okay. But yes, there is a general sort of advice that if you can take a decent center over a better shooting guard, then it’s a good idea to do that. But! I will remind you that I won the DroppingDimes Experts League with Nick Collison and Al Horford as my centers. Granted, both of them played well above where I drafted them, but that is the kind of example that demonstrates my reasoning for taking Kobe over Amare.
And Patrick’s response:
Not only can positional adjustments be done, but I took a look at it a while back (I forgive you for not reading the posts, I could barely find them myself). The first post is the most detailed and describes my methodology.
http://givemetherock.com/2006/08/13/702/The second one is basically an update using data through 2007.
http://givemetherock.com/2007/09/09/positional-value/
My apologies for all the crazy symbols in the posts, things got screwy when we did our latest Wordpress update. I don’t know if you have player ratings in addition to your rankings, but I basically took the GMTR player ratings and created an average positional rating using the top 100 players (40 guards, 40 forwards, 20 centers). And the top 40 guards had an average rating of about 1, while the top 20 centers were around -0.8. I then used these “positional adjustments” on all the individual player ratings, so guards got 1 point subtracted from the ratings (Kobe goes from 6 to 5, for example) and centers get 0.8 points added to their rating (Amare goes from 5.2 to 6). After that adjustment, Amare does indeed have slightly more value than Kobe.
I’m sure there are other and more accurate ways of accounting for positions. My adjustment was fixed, while in reality dropping from Amare to the second best center is a much bigger fall than going from Kobe to say, Baron Davis. You may want to compare the drop in player values by position as you move down your list.
To expand a little more: I personally prefer to take who I feel is the better player regardless of position for at least the first 3 rounds of a normal sized draft. If I feel I’m lacking at a position later in the draft, I’ll fill that with a player who I think will be playing above where I actually take them. I realize this explanation involved a lot more handwaving than something like Patrick’s calculations, but at the same time, making those calculations only tells you where a player was ranked, and can’t predict future statistics, let alone injuries and trades.
So, the simple answer is: yes, you need to take depth at a position into account when drafting. But, as with any advice in a fantasy draft, don’t follow it blindly. There are too many such ideas, and the key to fantasy drafts is learning as many of those ideas as possible, and synthesizing them into a strategy with which you are comfortable, and which proves successful. The latter you’ll only know after the season is over.
Depth at position is actually a fairly advanced concept from the way I look at fantasy basketball, so I’m glad to be able to talk about it now (during the “off-season”) when I feel like more advanced players will be searching for this kind of knowledge (as opposed to September and October when a lot of new people start looking for basic strategies and advice).
Subscribe via RSS